"David E. Wheeler" <david(at)kineticode(dot)com> writes:
> And yet they're DISTINCT FROM each other if either or both contain
> NULLs and some other values?
Well, that would depend on what the values were and in what columns...
> It seems to me that, to be consistent, it
> should be:
> select ROW(1, NULL) IS DISTINCT FROM ROW(2, NULL);
> ?column?
> ----------
> f
Dunno how you can possibly come to that conclusion. Those row
values are certainly distinct (according to both PG and the spec).
regards, tom lane